I am reading through the ENCODE papers, which is taking me well out of my comfort zone in terms of modern laboratory techniques. At the risk of asking a question which may well be thoroughly answered somewhere else on the internet, I was hoping for a brief explanation of nuclear transfection.

In Whiteld et all, they describe the direct assay of transcription factor binding site efficacy in human cells. To do this, they transfect a luciferase plasmid, presumably into the nucleus of the cell (because they subsequently observe that the luciferase plasmid is transcribed).

Reading through Materials and Methods and getting to the supplementary file, they disclosed their transfection protocol. The relevant reagants appear to be:

FuGene 6

a nonliposomal formulation designed to transfect plasmid DNA into a wide variety of cell lines with high efficiency and low toxicity.

(From Promega's website.)

Lipofectamine LTX

efficient reagent for plasmid delivery and protein expression

(From Invitrogen's website.)

PLUS reagent

PLUS™ Reagent is used in conjunction with transfection reagents, such as Lipofectamine™, to enhance transfection in adherent cell lines.

(From Invitrogen's website.)

Apparently, the protocol is to combine the DNA plasmid with transfection reagent, wait a few minutes, and then add in the cells, and incubate at 37ºC for 24 hours. That's all that is required to get the plasmid into the nucleus.

I have consulted Wikipedia and Strachan and Read's Human Molecular Genetics 4th edition, but been unable to answer my question about this technique.

I think I have a pretty clear understanding of how these reagents are able to get through the cell's lipid membrane, but I don't have a similar understanding of how the DNA then passes to the nucleus.

In Human Molecular Genetics 4th edition, I can find only one paragraph about plasmid transfer across a nuclear membrane, page 704: "Transport of plasmid DNA to the nucleus of non-dividing cells is very inefficient because the plasmid DNA often cannot enter nuclear membrane pores. Various methods can be used to faciliate nuclear entry such as conjugating specific DNA sequences or protein sequences (nuclear localization seuqences) that are known to facilitate nuclear entry, or compacting the DNA to a small enough size to pass through the nuclear pores."

But nothing in the materials and methods section of Whiteld et. all mentions any of these techniques. I'm sure they would have mentioned if their plasmid included a DNA conjugation sequence, so that can't be it. Perhaps these reagents implicitly include some sort of DNA compaction polycation, such as PEG-CK30?

Isn't the cytoplasm an extremely hostile place for DNA plasmids? Shouldn't they just flail around in the cytoplasm until they are digested?

Or perhaps these vectors sneak into the nucleus during mitosis? (This is a theory that just occurred to me when contemplating the "non-dividing" qualification in the Strachan quote.)

Since the question isn't addressed in the paper, I can only assume the answer is elementary and something that a professional cell biologist would be expected to know. I've run out of obvious next steps for finding the answer, so I beg pardon in advance for asking it.


3 Answers 3


My attempt to find an answer has suggested that no-one knows how the DNA gets into the nucleus.

This fairly recent paper reports attempts to track the pathway of DNA entry and transfer to the nucleus.

Le Bihan et al. (2010) Probing the in vitro mechanism of action of cationic lipid/DNA lipoplexes at a nanometric scale. Nucl. Acids Res. 39:1595-1609

The technique used was transmission electron microscopy. Complexes of plasmid DNA + cationic lipids (lipoplexes) were labelled with silica-based nanoparticles to allow visualization.

The authors describe a transfection pathway with five steps:

(i) binding of the lipoplexes to the cell surface;
(ii) entry of the lipoplexes into the cells mainly via endocytosis (a direct fusion with cell membrane cannot be excluded);
(iii) DNA release into the cytosol;
(iv) transport through the cytosol; and
(v) entry of DNA into the nucleus and nuclear transcription.

The authors were able to follow each of the first four steps but:

our data show nanoparticles close to, but not within, the nucleus.

They speculate about what this might mean, but they do not refer to any previously published data. My conclusion is that the answer to your question is, as I said at the beginning of my answer: no-one knows how the transfected DNA enters the nucleus.

  • $\begingroup$ Thanks Alan! I'm upvoting this for the nice article summary, but leaving the question unanswered in case someone else can comment specifically on Whiteld or shed some more light. $\endgroup$
    – masonk
    Commented Sep 14, 2012 at 16:05
  • $\begingroup$ The paper describes silicon nanoparticles in the 10 - 15 nm range, so it's not surprising they don't enter the nucleus, the nuclear pore complex won't let particles larger than about 5 nm through by passive diffusion and active transport requires a nuclear localizing peptide. Once the nanoparticles are attached to DNA liposomes, the sizes are even larger, maybe larger than the 39 nm maximum limit for nuclear pore complexes. $\endgroup$
    – user137
    Commented Jul 15, 2015 at 16:52

Zabner et al (J. Biol. Chem., 1995, 270, 18997-19007) monitored the lipid-DNA complex formation and uptake which was found to be predominantly by endocytosis. Dot blot experiment shows that cells (50% population) took up 60% of the DNA. The DNA-protein complex was observed to be deposited in the perinuclear cytoplasm and formed a series of regularly packed tubules. They also verified that the endosomes carrying the complex did not fuse with lysosome.

Some of the DNA escapes the endosome (which is a known mechanism of increasing the efficiency of endocytosis based transfections) and might be taken up by the nucleus. When they compared expression of a reporter gene upon direct injection into the nucleus and upon direct injection into the cytoplasm, no expression was observed in the latter (acc. to Capecchi (Cell, 1980) the ratio of expression is around 1000:1. PDF available on Utah Uni Website). This means that the DNA uptake by the nucleus from cytoplasm is very inefficient.
Lower amount of lipids in protein-DNA complex increases the efficiency.

Your question has been partly answered. How does all that DNA cell takes up gets inside the nucleus? It mostly doesn't. It accumulates around the nucleus. But some of the free DNA does get taken up by the nucleus. This, probably, was the starting point for Dean's group.

Vaughan & Dean (Mol.Ther., 2006, 13(2) 422-428) have proposed that the plasmids get transported into the nucleus with the help of microtubule associated motor protein Dynein. They observed that disrupting the microtubule network functionality inhibited expression. When the plasmids were injected directly into the nuclei of cells, this did not have an effect. Now, microtubules transport stuff with the help of motor proteins out of which Dynein transports stuff towards the nucleus. It was observed that inhibiting Dynein action reduced the expression. The mechanism of how dynein binds to plasmid is unknown. (I've worked with dynein; it's complicated when it comes to binding mechanisms. The whole negatively charged N-terminal is disordered.)
If you read the discussion of the paper, It'd become apparent that the process is chaotic. The more the dispersion of DNA in cytoplasm around the nucleus, the more is it's uptake by the nucleus.

There is no Nuclear localization signal involved here. Mesika et al (Human Gene Ther., 2005) have increased the efficiency of plasmid expression by binding the plasmids to a protein with NLS. This is not the natural process. Vaughan and Dean have speculated that the DNA targeting sequence (e.g. SV40, CMV etc.) binds to proteins that have NLS or have adapters to bind dynein(like, Transcription factors) which might help the plasmid uptake by the nucleus. This can also explain the sequence based differential uptake of plasmids. None of these mechanisms have been experimentally verified.

[I have also been wondering about this. I bugged a lot of seniors and scoured the net for the answers. And then I found this question. Thank you for asking]

  • 1
    $\begingroup$ Welcome to biology SE - you should summarise the paper a little more than you have so that the answer can stand-alone (e.g. what if the link goes dead), otherwise, it should just be a comment $\endgroup$
    – rg255
    Commented Jul 15, 2015 at 16:33
  • $\begingroup$ This is one of David Dean's papers, they're not bad, but have been hard to replicate. He basically proposes that transcription factor binding sites on plasmid DNA binds to some transcription factor in the cytoplasm, then the NLS peptide on the transcription factor binds the whole protein-DNA complex to importin and drags it to the nucleus. $\endgroup$
    – user137
    Commented Jul 15, 2015 at 16:56
  • $\begingroup$ Dean likes to use the SV40 promoter region, which is 72 bases long and contains several transcription factor binding sites. I'm not sure if anyone's tried it, but it should be possible to just make a 100 base oligo that contains the SV40 sequence and fluorescently label it. See if that goes to the nucleus. $\endgroup$
    – user137
    Commented Jul 16, 2015 at 16:29

This is a question that keeps coming up now and then when PhDs or post-docs discuss transfection. A probable answer for which I do not have supporting literature is this: If you have the transfection complex in the medium while the cells are dividing, it is possible that they can enter the nucleus following mitosis when the nuclear membrane is still forming, thus giving it access to the transciption mechanism in the nucleus.

  • $\begingroup$ Hello and welcome to the site. Is there no supporting literature, or do you just not have it at hand? If the latter, then you are encouraged to add citations. $\endgroup$
    – kmm
    Commented Oct 9, 2013 at 13:35
  • $\begingroup$ I don't want to add the citations myself, but this is most likely the case for rapidly dividing cells in culture. You really should find some supporting literature to improve this answer. $\endgroup$
    – user137
    Commented Jul 15, 2015 at 16:46
  • $\begingroup$ This could be one of the mechanisms, but does not explain the fact that the reagents above are useful for post-mitotic cells such as neurons. $\endgroup$
    – 243
    Commented Jul 16, 2015 at 15:16
  • $\begingroup$ @243 They may not be as useful as you think. What kind of transfection efficiencies are you seeing, and how do they compare with rapidly dividing cells using lipofectamine or PEI? $\endgroup$
    – user137
    Commented Jul 16, 2015 at 16:11
  • $\begingroup$ FuGene 6 and Lipofectamine LTX are used to transfect plasmids into post mitotic neurons primary cultured. I would not say efficiency is high, but acceptable for some experiments. For example: ncbi.nlm.nih.gov/pubmed/11716945 $\endgroup$
    – 243
    Commented Jul 16, 2015 at 23:41

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .